EP3876144A1 - Face identification method and face identification apparatus - Google Patents
Face identification method and face identification apparatus Download PDFInfo
- Publication number
- EP3876144A1 EP3876144A1 EP20193842.0A EP20193842A EP3876144A1 EP 3876144 A1 EP3876144 A1 EP 3876144A1 EP 20193842 A EP20193842 A EP 20193842A EP 3876144 A1 EP3876144 A1 EP 3876144A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- infrared light
- face identification
- data processing
- face
- processing circuit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000010586 diagram Methods 0.000 description 4
- OAWXYINGQXLWOE-UHFFFAOYSA-N (2-acetyloxybenzoyl) 2-acetyloxybenzoate Chemical compound CC(=O)OC1=CC=CC=C1C(=O)OC(=O)C1=CC=CC=C1OC(C)=O OAWXYINGQXLWOE-UHFFFAOYSA-N 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000004709 eyebrow Anatomy 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
- G01S17/08—Systems determining position data of a target for measuring distance only
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/141—Control of illumination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/76—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries based on eigen-space representations, e.g. from pose or different illumination conditions; Shape manifolds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Definitions
- This disclosure relates to face identification, in particular to a face identification method and a face identification apparatus.
- a face identification technology is to find a face in an image, and then compare the found face with contents of a database to compare whether the captured face matches an identity data in the database.
- the captured image may be a visible light image or an invisible light image.
- a common method is to digitize features of different parts of the face to generate a face eigenvector combination, and then the face eigenvector combination is stored into the database as the identity data, or compared with stored identity data in the database.
- an image camera continuously captures a visible or invisible streaming image.
- a central processing unit or other computing units continuously perform identification program instructions, analyze whether there is a face in the streaming image and further generate a face eigenvector combination when the face appears. That is to say, whether there is a user in front of the face identification apparatus, a face identification mechanism continuously performs identification work.
- the central processing unit of the face identification apparatus is continuously in a high-computing-load state. If a central processing unit with low computing capability is used, the high-computing-load state will make a whole system operate slowly, which affects operation of other tasks. However, if a central processing unit with high computing capability is adopted, the cost is too high, and meanwhile, a continuous high load also brings high power consumption. High heat brought by high power consumption also often makes the system enter into a low clock rate operation for cooling. The low clock operation further causes hysteresis of system operation.
- this disclosure discloses a face identification method and a face identification apparatus, which can duly start face identification program to reduce an average load of a system and power consumption of hardware.
- This disclosure discloses a face identification method, including: performing a distance detection to obtain a detected distance value; determining whether the detected distance value is smaller than a distance threshold; when the detected distance value is smaller than the distance threshold, determining a luminance corresponding to the detected distance value and emitting an infrared light according to the luminance; capturing an infrared light image and performing face identification to the infrared light image; and when the face identification is successful, performing a corresponding event.
- a range of the distance threshold is divided into a plurality of distance intervals, the luminance has a plurality of set values, and each of the distance intervals corresponds to one of the plurality of set values.
- the face identification includes: analyzing whether there is a face in the infrared light image, and correcting an image of the face; capturing a face eigenvector combination of the face; and comparing whether the face eigenvector combination matches one of a plurality of preset eigenvector combinations in an eigenface database.
- the face identification method further includes: when the detected distance value is smaller than the distance threshold, capturing a visible light image, and displaying the visible light image.
- the face identification method further includes: when the face identification is successful, displaying a successful identification prompt, wherein the successful identification prompt includes a user identity data.
- This disclosure further discloses a face identification apparatus, including a distance detector, an infrared light transmitter, an infrared light camera, and a data processing circuit.
- the distance detector is configured to obtain a detected distance value.
- the infrared light transmitter is configured to emit an infrared light.
- the infrared light camera is configured to capture an infrared light image.
- the data processing circuit is configured to determine whether the detected distance value is smaller than a distance threshold.
- the data processing circuit determines a luminance corresponding to the detected distance value, and drives the infrared light transmitter to emit the infrared light according to the luminance.
- the data processing circuit performs face identification to the infrared light image. When the face identification is successful, the data processing circuit performs a corresponding event.
- the face identification apparatus further includes a visible light camera, configured to capture a visible light image and transmit the visible light image to the data processing circuit; and a display panel, configured to be driven by the data processing circuit to display the visible light image.
- the face identification apparatus further includes a casing.
- the casing includes an internal space and a front surface.
- the distance detector is disposed on the front surface and configured to obtain the detected distance value in front of the front surface.
- the data processing circuit is disposed in the internal space.
- the infrared light transmitter, the infrared light camera, the visible light camera and the display panel are disposed on the front surface.
- the data processing circuit drives the display panel to display a successful identification prompt, and the successful identification prompt includes a user identity data.
- a range of the distance threshold is divided into a plurality of distance intervals, the luminance has a plurality of set values, and each of the distance intervals corresponds to one of the plurality of set values.
- the data processing circuit includes an computing unit, electrically connected to the infrared light camera; a driving circuit, connected to the computing unit and the distance detector, wherein the detected distance value is coded by the driving circuit and then transmitted to the computing unit; and a voltage regulator, connected to the driving circuit and the infrared light transmitter, and receiving a reference voltage from the driving circuit.
- the computing unit controls the driving circuit to switch the reference voltage according to the detected distance value, so as to regulate a magnitude of a driving voltage output to the infrared light transmitter by the voltage regulator, so that the luminance of the infrared light transmitter corresponds to the detected distance value.
- the face identification apparatus switches a face identification program to be in an idle state and being awakened and started at any time.
- the face identification apparatus determines whether the face identification program is awakened and started according to the detected distance value. Therefore, the face identification apparatus does not continuously perform the face identification program, so that continuous consumption of hardware computing resources into the face identification program is avoided. Therefore, this disclosure can effectively reduce an average computing load of the face identification apparatus and avoids unnecessary power consumption. Meanwhile, reducing the load of the system can avoid an over-temperature condition of the system to avoid the system entering into a low clock rate operation, so that the face identification program can be quickly completed.
- this disclosure utilizes the detected distance value to determine the set values of the luminance configured to illuminate, so as to guarantee that optimal image exposure and image capturing effects of the face can be obtained from far to near, which facilitates performing of the face identification program.
- a face identification apparatus 100 disclosed by a first embodiment of this disclosure is shown, and is configured to perform a face identification method.
- the face identification apparatus 100 includes a casing 110, a distance detector 120, a data processing circuit 130, one or more infrared light transmitters 140, an infrared light camera 150, a visible light camera 160 and a display panel 170.
- the casing 110 includes an internal space and a front surface 112.
- the distance detector 120 is disposed on the front surface 112 and configured to obtain a detected distance value D in front of the front surface 112.
- the casing 110 is usually fixed to a wall surface or a bracket and faces a detecting area by the front surface 112.
- the data processing circuit 130 is disposed in the internal space of the casing 110.
- the distance detector 120 is electrically connected to the data processing circuit 130 and transmits the detected distance value D to the data processing circuit 130.
- the data processing circuit 130 determines a distance threshold T and determines whether the detected distance value D is smaller than the distance threshold T.
- the detected distance value D is fixed value set to be larger than the distance threshold T. For example, since a distance of a background object capable of being detected by the distance detector 120, or a distance of the background object exceeds the maximum detection range of the distance detector 120, an effective detected distance value D cannot be obtained.
- the infrared light transmitters 140 are disposed on the front surface 112 and electrically connected to the data processing circuit 130.
- the infrared light transmitters 140 are configured to emit an infrared light in front of the front surface 112.
- the data processing circuit 130 determines a luminance corresponding to the detected distance value D and drives the infrared light transmitters 140 to emit the infrared light according to the corresponding luminance.
- the luminance may be light emission power (watt) of the infrared light transmitters 140; for example, the infrared light transmitters 140 emit the infrared light at a specified power; or the luminance may be a magnitude of driving voltage for driving the infrared light transmitters 140.
- the data processing circuit 130 may determine driving voltage values corresponding to different detected distance values D within a driving voltage range of the infrared light transmitters 140.
- the infrared light transmitters 140 are driven by a pulse current.
- the luminance may be a width of the pulse current or a duty cycle length of the pulse current.
- the data processing circuit 130 may determine the width or a duty cycle of the pulse current corresponding to different detected distance values D within a working frequency range of the infrared light transmitters 140. The larger the detected distance value D is, the larger the corresponding luminance is. Otherwise, the smaller the detected distance value D is, the smaller the corresponding luminance is.
- the data processing circuit 130 further divides a range of the distance threshold T into two or more distance intervals D1, D2, D3, and determines the luminance to have a plurality of set values I1, I2, I3.
- the data processing circuit 130 determines the corresponding one of the set values I1, I2, I3 for each of the distance intervals D1, D2, D3, so as to drive the infrared light transmitters 140 to emit the infrared light.
- the distance detector 120 is still capable of obtaining a detected distance value D in a distance interval D4 not larger than the maximum detection distance, and the data processing circuit 130 is also capable of determining a corresponding set value I4 for the distance interval D4.
- the set value I4 may be the maximum luminance within a capability of the infrared light transmitters 140.
- the data processing circuit 130 drives the infrared light transmitters 140 to emit the Iinfrared light at the corresponding one of the set values I1, I2, I3.
- an infrared light irradiance (W/m2) of a surface of an object can be maintained within an appropriate numerical value range when the object is in the distance intervals D1, D2, D3.
- the infrared light camera 150 is disposed on the front surface 112 and electrically connected to the data processing circuit 130.
- the infrared light camera 150 is configured to capture an infrared light image Pir in front of the front surface 112 of the casing 110 and transmit the infrared light image Pir to the data processing circuit 130.
- the infrared light image Pir is mainly formed after the infrared light emitted by the infrared light transmitters 140 is reflected by the object.
- the data processing circuit 130 continuously determines whether the detected distance value D is smaller than the distance threshold T. When the detected distance value D is smaller than the distance threshold T, the data processing circuit 130 determines the luminance corresponding to the detected distance value D and drives the infrared light transmitters 140 to emit the infrared light according to the luminance. The data processing circuit 130 starts the infrared light camera 150 at the same time to capture the infrared light image Pir. The data processing circuit 130 performs face identification according to the infrared light image Pir. The data processing circuit 130 analyzes whether there is a face in the infrared light image Pir, corrects an image of the face and then captures a face eigenvector combination of the face.
- the face eigenvector combination is generated by respectively taking out eigenvectors by utilizing facial feature points (such as eyebrow, eyes, nose and lips) of the face as anchor points.
- the data processing circuit 130 After capturing the face eigenvector combination, the data processing circuit 130 compares whether the face eigenvector combination matches one of a plurality of preset eigenvector combinations in an eigenface database. When the face eigenvector combination matches one of the plurality of preset eigenvector combinations, the face identification is successful, and the data processing circuit 130 performs a corresponding event.
- the visible light camera 160 and the display panel 170 are electrically connected to the data processing circuit 130.
- the visible light camera 160 and the display panel 170 are disposed on the front surface 112 of the casing 110.
- the visible light camera 160 is configured to capture a visible light image Pv in front of the front surface 112 of the casing 110 and transmit the visible light image Pv to the data processing circuit 130, so that the data processing circuit 130 drives the display panel 170 to display the visible light image Pv. Therefore, a user can confirm and regulate a self-position through display of the display panel 170 so as to guarantee that the self-position and a posture can make the infrared light camera 150 be capable of capturing his/her face.
- the foregoing corresponding event may be, but is not limited to, door lock opening, attendance recording, access control recording, book borrowing recording, etc.
- a gate is arranged in a workplace and controlled by the data processing circuit 130.
- the data processing circuit 130 may record office hours of the user, opening the gate, etc. as the corresponding event.
- the foregoing corresponding event performed by the data processing circuit 130 may further include the work that the data processing circuit 130 drives the display panel 170 to display a successful identification prompt.
- the successful identification prompt may include displaying a user identity data.
- this disclosure further discloses a face identification method.
- the data processing circuit 130 enables the distance detector 120 to perform a distance detection so as to obtain the detected distance value D, as shown in Step 110.
- the data processing circuit 130 loads a face identification program but switches the face identification program to be in an idle state. That is, the face identification program is not enabled and is only in a state of being able to be started at any time.
- the face identification apparatus 100 at the moment will not perform the face identification so as to avoid computing resource consumption caused by the data processing circuit 130 continuously performing a high-load computing.
- the distance detector 120 continuously performs the distance detection, and continuously transmits the detected distance value D back to the data processing circuit 130, as shown in Step 120.
- the data processing circuit 130 determines a distance threshold T, and the data processing circuit 130 determines whether the detected distance value D is smaller than the distance threshold T, as shown in Step 130.
- Step 120 is repeated, and the data processing circuit 130 receives new detected distance value D transmitted back.
- the distance detector 120 detects the user, so that the corresponding detected distance value D is obtained according to a distance between the user and the distance detector 120.
- the data processing circuit 130 enables the face identification program, as shown in Step 140.
- enabling the face identification program includes that the data processing circuit 130 determines the luminance corresponding to the detected distance value D and drives the infrared light transmitters 140 to emit the infrared light according to the corresponding luminance, as shown in Step 141 and Step 142. Then, the data processing circuit 130 starts the infrared light camera 150 to start to capture the infrared light image Pir and transmit the infrared light image back, as shown in Step 143.
- the corresponding luminance is determined to emit the infrared light so as to guarantee that the infrared light image Pir can have good image exposure and a clear image capturing effect.
- the specific implementation is shown in Fig. 3 , the range of the distance threshold T is divided into the two or more distance intervals D1, D2, D3, the luminance has the plurality of set values I1, I2, I3, and each of the plurality of distance intervals D1, D2, D3 corresponds to one of the plurality of set values I1, I2, I3.
- the data processing circuit 130 does not perform the face identification but is still capable of driving the infrared light transmitters 140 to emit the infrared light at the set value 14, so that the infrared light transmitters 140 maintain the set value I4 of the maximum luminance within the capability of the luminance to emit the infrared light, and surroundings are illuminated by the infrared light.
- the infrared light transmitters 140 take the maximum set value I3 as the luminance firstly to emit the infrared light, so that the face can still be fully illuminated at a long distance, and it is guaranteed that the infrared light image Pir is sufficiently exposed and is clear enough.
- the data processing circuit 130 determines the corresponding luminance as the set value I2 according to the distance interval D2, so that the infrared light image Pir can be moderately exposed and the face is clearly presented in the infrared light image Pir.
- the infrared light transmitters 140 will take the smaller set value I1 as the luminance to emit the infrared light, so that the infrared light image Pir is moderately exposed and the face is clear enough. But if the user moves to change the distance to increase the detected distance value D to be in the distance interval D3 or the distance interval D2, the luminance of the set value I1 will be insufficient, which results in that the infrared light image Pir is underexposed and is too dark, and consequently, the features of the cannot be clearly presented infrared light image Pir. At the moment, the data processing circuit 130 determines the corresponding luminance to be increased as the set value I3 according to the distance interval D3, so that the face in the infrared light image Pir can be presented clearly.
- the data processing circuit 130 starts the visible light camera 160 to capture the visible light image Pv and drives the display panel 170 to display the visible light image Pv, as shown in Step 144.
- the user can regulate the self-position and the posture according to display of the display panel 170 so as to guarantee that the face of the user can be captured by the infrared light camera 150.
- the foregoing infrared light transmitters 140 may normally emit the infrared light at the maximum luminance such as the set value I4 within the capability to perform infrared light illumination around.
- the infrared light camera 150 continuously captures an infrared light image Pir stream and stores the stream to the data processing circuit 130. But the data processing circuit 130 at the moment does not perform the face identification and only records a video.
- the visible light camera 160 may also normally capture a visible light image Pv stream.
- the data processing circuit 130 records the video, but does not drive the display panel 170 to display the video.
- the data processing circuit 130 performs the face identification according to the infrared light image Pir and compare whether the face eigenvector combination of the face of the infrared light image Pir matches one of the plurality of preset eigenvector combinations in the eigenface database, as shown in Step 150.
- the above-mentioned plurality means including two or more.
- the face identification is successful, and the data processing circuit 130 performs the corresponding event, as shown in Step 160.
- the foregoing corresponding event may be, but is not limited to, door lock opening, attendance recording, access control recording, book borrowing recording, etc.
- the face identification program is normally in the idle state, so that hardware of the face identification apparatus 100 does not continuously maintain a high computing load due to continuously performing the face identification program.
- a usual load of the face identification apparatus 100 can be reduced, and power consumption is reduced.
- the data processing circuit 130 may be implemented by hardware or a combination of the hardware and software.
- the data processing circuit 130 may further include a system logic chipset, a central processing unit, a non-volatile memory module, a volatile memory module and a graphic processing unit.
- the non-volatile memory module is configured to store program instructions.
- the central processing unit may load the instructions from a non-volatile memory to a volatile memory module to be performed so as to start the face identification program and switch the face identification program to be in the idle state.
- a face identification apparatus 100 disclosed by a second embodiment of this disclosure is shown, and is configured to provide an example of another data processing circuit 130.
- the data processing circuit 130 may be divided into three parts including a computing unit 132, a driving circuit 134 of a distance detector 120 and a voltage regulator 136.
- the computing unit 132 includes components such as a central processing unit (CPU) and a system logic chipset and is configured to perform face identification.
- the computing unit 132 may be connected to an infrared light camera 150 and a visible light camera 160 through a USB bus (or other interfaces such as an MIPI-CSI).
- the computing unit 132 is connected to the driving circuit 134 of the distance detector 120 through an I2C interface (or other interfaces such as an SMBus and a USB). Meanwhile, the driving circuit 134 may be connected to the voltage regulator 136 through two output pins such as general input and output pins GPIO-1 and GPIO-2.
- the general input and output pins GPIO-1 and GPIO-2 and a base voltage V are in parallel connection with the voltage regulator 136 respectively through resistors, so that a reference voltage Vref is regulated and output to the voltage regulator 136 through the general input and output pins GPIO-1 and GPIO-2.
- the voltage regulator 136 is connected to a power source Vdd, such as a power source with +5V voltage, and outputs an output voltage Vout according to the reference voltage Vref to regulate a magnitude of a driving voltage V-LED of infrared light transmitters 140, so that one of set values I1, I2, I3 and I4 is selected to change a luminance.
- the distance detector 120 may be an IR distance detector, a distance detector and a time of flight (TOF) detector.
- a detected distance value D detected by the distance detector 120 is coded by the driving circuit 134 and then continuously transmitted back to the CPU of the computing unit 132.
- the data processing circuit 130 does not perform face identification and does not switch the set values I1, I2, I3 and I4 of the luminance of the infrared light transmitters 140 either.
- the computing unit 132 controls the driving circuit 134 of the distance detector 120 to make the two general input and output pins GPIO-1 and GPIO-2 be both maintained in an open drain state.
- the infrared light transmitters 140 maintain the set value I4 of the maximum luminance within a capability as the luminance to emit an infrared light, and surroundings are illuminated by the infrared light.
- the computing unit 132 controls the driving circuit 134 of the distance detector 120 to make the two general input and output pins GPIO-1 and GPIO-2 be both maintained in a low level state.
- the computing unit 132 controls the driving circuit 134 of the distance detector 120 through I2C, so that the general input and output pin GPIO-1 is switched to be at a low level, and the general input and output pin GPIO-2 is switched to be in an open drain state.
- the computing unit 132 controls the driving circuit 134 of the distance detector 120 through the I2C, so that the general input and output pin GPIO-1 is switched to be in an open drain state, and the general input and output pin GPIO-2 is switched to be at a low level.
- the three distance intervals D1, D2, D3 corresponding to the three set values I1, I2, I3 of the luminance are only examples, and the corresponding number thereof is not limited to three and may be two or four or more. Meanwhile, at least one distance interval (such as the distance interval D4) can still be set above the distance threshold T, and the distance interval D4 also corresponds to one set value 14.
- the face identification apparatus 100 switches the face identification program to be in the state of being idle and being awakened and started at any time.
- the face identification apparatus 100 determines whether the face identification program is awakened and started according to the detected distance value D. Therefore, the face identification apparatus 100 does not continuously perform the face identification program, so that continuous consumption of hardware computing resources into the face identification program is avoided. Therefore, this disclosure can effectively reduce an average computing load of the face identification apparatus 100 and avoids unnecessary power consumption. Meanwhile, reducing the load of the system can avoid an over-temperature condition of the system to avoid the system entering into a low clock rate operation, so that the face identification program can be quickly completed.
- this disclosure utilizes the detected distance value D to determine the set values I1, I2, I3 of the luminance configured to illuminate, so as to guarantee that the optimal image exposure and image capturing effects of the face can be obtained from far to near, which facilitates performing of the face identification.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Electromagnetism (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Networks & Wireless Communication (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Collating Specific Patterns (AREA)
- Image Input (AREA)
- Image Analysis (AREA)
- Studio Devices (AREA)
- Measurement Of Optical Distance (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
A face identification method includes performing a distance detection to obtain a detected distance value (D); determining whether the detected distance value (D) is smaller than a distance threshold (T); when the detected distance value (D) is smaller than the distance threshold (T), determining a luminance corresponding to the detected distance value (D) and emitting an infrared light according to the luminance; capturing an infrared light image (Pir) and performing face identification to the infrared light image (Pir); and when the face identification is successful, performing a corresponding event. A face identification apparatus (100) configured to perform the face identification method is further provided.
Description
- This disclosure relates to face identification, in particular to a face identification method and a face identification apparatus.
- A face identification technology is to find a face in an image, and then compare the found face with contents of a database to compare whether the captured face matches an identity data in the database. The captured image may be a visible light image or an invisible light image. A common method is to digitize features of different parts of the face to generate a face eigenvector combination, and then the face eigenvector combination is stored into the database as the identity data, or compared with stored identity data in the database.
- As for a current face identification apparatus, an image camera continuously captures a visible or invisible streaming image. A central processing unit or other computing units continuously perform identification program instructions, analyze whether there is a face in the streaming image and further generate a face eigenvector combination when the face appears. That is to say, whether there is a user in front of the face identification apparatus, a face identification mechanism continuously performs identification work.
- Under the above situation, the central processing unit of the face identification apparatus is continuously in a high-computing-load state. If a central processing unit with low computing capability is used, the high-computing-load state will make a whole system operate slowly, which affects operation of other tasks. However, if a central processing unit with high computing capability is adopted, the cost is too high, and meanwhile, a continuous high load also brings high power consumption. High heat brought by high power consumption also often makes the system enter into a low clock rate operation for cooling. The low clock operation further causes hysteresis of system operation.
- In view of the above problem, this disclosure discloses a face identification method and a face identification apparatus, which can duly start face identification program to reduce an average load of a system and power consumption of hardware.
- This disclosure discloses a face identification method, including: performing a distance detection to obtain a detected distance value; determining whether the detected distance value is smaller than a distance threshold; when the detected distance value is smaller than the distance threshold, determining a luminance corresponding to the detected distance value and emitting an infrared light according to the luminance; capturing an infrared light image and performing face identification to the infrared light image; and when the face identification is successful, performing a corresponding event.
- In at least one embodiment, a range of the distance threshold is divided into a plurality of distance intervals, the luminance has a plurality of set values, and each of the distance intervals corresponds to one of the plurality of set values.
- In at least one embodiment, the face identification includes: analyzing whether there is a face in the infrared light image, and correcting an image of the face; capturing a face eigenvector combination of the face; and comparing whether the face eigenvector combination matches one of a plurality of preset eigenvector combinations in an eigenface database.
- In at least one embodiment, the face identification method further includes: when the detected distance value is smaller than the distance threshold, capturing a visible light image, and displaying the visible light image.
- In at least one embodiment, the face identification method further includes: when the face identification is successful, displaying a successful identification prompt, wherein the successful identification prompt includes a user identity data.
- This disclosure further discloses a face identification apparatus, including a distance detector, an infrared light transmitter, an infrared light camera, and a data processing circuit. The distance detector is configured to obtain a detected distance value. The infrared light transmitter is configured to emit an infrared light. The infrared light camera is configured to capture an infrared light image. The data processing circuit is configured to determine whether the detected distance value is smaller than a distance threshold.
- When the detected distance value is smaller than the distance threshold, the data processing circuit determines a luminance corresponding to the detected distance value, and drives the infrared light transmitter to emit the infrared light according to the luminance. The data processing circuit performs face identification to the infrared light image. When the face identification is successful, the data processing circuit performs a corresponding event.
- In at least one embodiment, the face identification apparatus further includes a visible light camera, configured to capture a visible light image and transmit the visible light image to the data processing circuit; and a display panel, configured to be driven by the data processing circuit to display the visible light image.
- In at least one embodiment, the face identification apparatus further includes a casing. The casing includes an internal space and a front surface. The distance detector is disposed on the front surface and configured to obtain the detected distance value in front of the front surface. The data processing circuit is disposed in the internal space. The infrared light transmitter, the infrared light camera, the visible light camera and the display panel are disposed on the front surface.
- In at least one embodiment, when the face identification is successful, the data processing circuit drives the display panel to display a successful identification prompt, and the successful identification prompt includes a user identity data.
- In at least one embodiment, a range of the distance threshold is divided into a plurality of distance intervals, the luminance has a plurality of set values, and each of the distance intervals corresponds to one of the plurality of set values.
- In at least one embodiment, the data processing circuit includes an computing unit, electrically connected to the infrared light camera; a driving circuit, connected to the computing unit and the distance detector, wherein the detected distance value is coded by the driving circuit and then transmitted to the computing unit; and a voltage regulator, connected to the driving circuit and the infrared light transmitter, and receiving a reference voltage from the driving circuit. The computing unit controls the driving circuit to switch the reference voltage according to the detected distance value, so as to regulate a magnitude of a driving voltage output to the infrared light transmitter by the voltage regulator, so that the luminance of the infrared light transmitter corresponds to the detected distance value.
- In one or more embodiments of this disclosure, the face identification apparatus switches a face identification program to be in an idle state and being awakened and started at any time. The face identification apparatus determines whether the face identification program is awakened and started according to the detected distance value. Therefore, the face identification apparatus does not continuously perform the face identification program, so that continuous consumption of hardware computing resources into the face identification program is avoided. Therefore, this disclosure can effectively reduce an average computing load of the face identification apparatus and avoids unnecessary power consumption. Meanwhile, reducing the load of the system can avoid an over-temperature condition of the system to avoid the system entering into a low clock rate operation, so that the face identification program can be quickly completed.
- In addition, in at least one embodiment, this disclosure utilizes the detected distance value to determine the set values of the luminance configured to illuminate, so as to guarantee that optimal image exposure and image capturing effects of the face can be obtained from far to near, which facilitates performing of the face identification program.
- This disclosure will become more fully understood from the detailed description given herein below for illustration only, and thus not limitative of this disclosure, wherein:
-
Fig. 1 is a front view according to a first embodiment of this disclosure; -
Fig. 2 is a circuit block diagram of the first embodiment of this disclosure; -
Fig. 3 is a corresponding relation diagram of a detected distance value and a luminance in the first embodiment of this disclosure; -
Fig. 4 is a flow chart of a method of the first embodiment of this disclosure; -
Fig. 5 is another flow chart of the method of the first embodiment of this disclosure; -
Fig. 6 andFig. 7 are schematic diagrams of switching of the detected distance value and the luminance; and -
Fig. 8 is a circuit block diagram according to a second embodiment of this disclosure. - Referring to
Fig. 1 andFig. 2 , aface identification apparatus 100 disclosed by a first embodiment of this disclosure is shown, and is configured to perform a face identification method. - As show in
Fig. 1 andFig. 2 , theface identification apparatus 100 includes acasing 110, adistance detector 120, adata processing circuit 130, one or moreinfrared light transmitters 140, aninfrared light camera 150, avisible light camera 160 and adisplay panel 170. - As show in
Fig. 1 andFig. 2 , thecasing 110 includes an internal space and afront surface 112. Thedistance detector 120 is disposed on thefront surface 112 and configured to obtain a detected distance value D in front of thefront surface 112. Thecasing 110 is usually fixed to a wall surface or a bracket and faces a detecting area by thefront surface 112. - As shown in
Fig. 2 andFig. 3 , thedata processing circuit 130 is disposed in the internal space of thecasing 110. Thedistance detector 120 is electrically connected to thedata processing circuit 130 and transmits the detected distance value D to thedata processing circuit 130. Thedata processing circuit 130 determines a distance threshold T and determines whether the detected distance value D is smaller than the distance threshold T. When no object approaches in the detecting area, for example, when no human body or face enters the detecting area, the detected distance value D is fixed value set to be larger than the distance threshold T. For example, since a distance of a background object capable of being detected by thedistance detector 120, or a distance of the background object exceeds the maximum detection range of thedistance detector 120, an effective detected distance value D cannot be obtained. - As shown in
Fig. 1 ,Fig. 2 andFig. 3 , theinfrared light transmitters 140 are disposed on thefront surface 112 and electrically connected to thedata processing circuit 130. Theinfrared light transmitters 140 are configured to emit an infrared light in front of thefront surface 112. When the detected distance value D is smaller than the distance threshold T, thedata processing circuit 130 determines a luminance corresponding to the detected distance value D and drives the infraredlight transmitters 140 to emit the infrared light according to the corresponding luminance. In one specific embodiment, the luminance may be light emission power (watt) of the infraredlight transmitters 140; for example, the infraredlight transmitters 140 emit the infrared light at a specified power; or the luminance may be a magnitude of driving voltage for driving the infraredlight transmitters 140. Thedata processing circuit 130 may determine driving voltage values corresponding to different detected distance values D within a driving voltage range of the infraredlight transmitters 140. In one specific embodiment, the infraredlight transmitters 140 are driven by a pulse current. The luminance may be a width of the pulse current or a duty cycle length of the pulse current. Thedata processing circuit 130 may determine the width or a duty cycle of the pulse current corresponding to different detected distance values D within a working frequency range of the infraredlight transmitters 140. The larger the detected distance value D is, the larger the corresponding luminance is. Otherwise, the smaller the detected distance value D is, the smaller the corresponding luminance is. - As shown in
Fig. 3 , in one specific embodiment, thedata processing circuit 130 further divides a range of the distance threshold T into two or more distance intervals D1, D2, D3, and determines the luminance to have a plurality of set values I1, I2, I3. Thedata processing circuit 130 determines the corresponding one of the set values I1, I2, I3 for each of the distance intervals D1, D2, D3, so as to drive the infraredlight transmitters 140 to emit the infrared light. Above the distance threshold T, thedistance detector 120 is still capable of obtaining a detected distance value D in a distance interval D4 not larger than the maximum detection distance, and thedata processing circuit 130 is also capable of determining a corresponding set value I4 for the distance interval D4. The set value I4 may be the maximum luminance within a capability of the infraredlight transmitters 140. - Thus, in the above specific embodiment, the larger average distances of the distance intervals D1, D2, D3 are, the larger the corresponding set values I1, I2, I3 are. Otherwise, the smaller the average distances of the distance intervals D1, D2, D3 are, the smaller the corresponding set values I1, I2, I3 are. Thus, when the detected distance value D falls in one of the plurality of distance intervals D1, D2, D3, the
data processing circuit 130 drives the infraredlight transmitters 140 to emit the Iinfrared light at the corresponding one of the set values I1, I2, I3. According to corresponding switching of the set values I1, I2, I3 and the distance intervals D1, D2, D3, an infrared light irradiance (W/m2) of a surface of an object can be maintained within an appropriate numerical value range when the object is in the distance intervals D1, D2, D3. - As shown in
Fig. 1 andFig. 2 , the infraredlight camera 150 is disposed on thefront surface 112 and electrically connected to thedata processing circuit 130. The infraredlight camera 150 is configured to capture an infrared light image Pir in front of thefront surface 112 of thecasing 110 and transmit the infrared light image Pir to thedata processing circuit 130. The infrared light image Pir is mainly formed after the infrared light emitted by the infraredlight transmitters 140 is reflected by the object. - The
data processing circuit 130 continuously determines whether the detected distance value D is smaller than the distance threshold T. When the detected distance value D is smaller than the distance threshold T, thedata processing circuit 130 determines the luminance corresponding to the detected distance value D and drives the infraredlight transmitters 140 to emit the infrared light according to the luminance. Thedata processing circuit 130 starts the infraredlight camera 150 at the same time to capture the infrared light image Pir. Thedata processing circuit 130 performs face identification according to the infrared light image Pir. Thedata processing circuit 130 analyzes whether there is a face in the infrared light image Pir, corrects an image of the face and then captures a face eigenvector combination of the face. The face eigenvector combination is generated by respectively taking out eigenvectors by utilizing facial feature points (such as eyebrow, eyes, nose and lips) of the face as anchor points. After capturing the face eigenvector combination, thedata processing circuit 130 compares whether the face eigenvector combination matches one of a plurality of preset eigenvector combinations in an eigenface database. When the face eigenvector combination matches one of the plurality of preset eigenvector combinations, the face identification is successful, and thedata processing circuit 130 performs a corresponding event. - As shown in
Fig. 1 andFig. 2 , the visiblelight camera 160 and thedisplay panel 170 are electrically connected to thedata processing circuit 130. The visiblelight camera 160 and thedisplay panel 170 are disposed on thefront surface 112 of thecasing 110. - The visible
light camera 160 is configured to capture a visible light image Pv in front of thefront surface 112 of thecasing 110 and transmit the visible light image Pv to thedata processing circuit 130, so that thedata processing circuit 130 drives thedisplay panel 170 to display the visible light image Pv. Therefore, a user can confirm and regulate a self-position through display of thedisplay panel 170 so as to guarantee that the self-position and a posture can make the infraredlight camera 150 be capable of capturing his/her face. - The foregoing corresponding event may be, but is not limited to, door lock opening, attendance recording, access control recording, book borrowing recording, etc. For example, a gate is arranged in a workplace and controlled by the
data processing circuit 130. When the face identification is successful, thedata processing circuit 130 may record office hours of the user, opening the gate, etc. as the corresponding event. The foregoing corresponding event performed by thedata processing circuit 130 may further include the work that thedata processing circuit 130 drives thedisplay panel 170 to display a successful identification prompt. The successful identification prompt may include displaying a user identity data. - As shown in
Fig. 4 , based on the aboveface identification apparatus 100, this disclosure further discloses a face identification method. - As shown in
Fig. 4 , after theface identification apparatus 100 is started, thedata processing circuit 130 enables thedistance detector 120 to perform a distance detection so as to obtain the detected distance value D, as shown inStep 110. Thedata processing circuit 130 loads a face identification program but switches the face identification program to be in an idle state. That is, the face identification program is not enabled and is only in a state of being able to be started at any time. Theface identification apparatus 100 at the moment will not perform the face identification so as to avoid computing resource consumption caused by thedata processing circuit 130 continuously performing a high-load computing. - As shown in
Fig. 4 , thedistance detector 120 continuously performs the distance detection, and continuously transmits the detected distance value D back to thedata processing circuit 130, as shown inStep 120. Thedata processing circuit 130 determines a distance threshold T, and thedata processing circuit 130 determines whether the detected distance value D is smaller than the distance threshold T, as shown inStep 130. When the detected distance value D is larger than the distance threshold T,Step 120 is repeated, and thedata processing circuit 130 receives new detected distance value D transmitted back. - When the user appears in front of the
front surface 112 and enters the detecting area, thedistance detector 120 detects the user, so that the corresponding detected distance value D is obtained according to a distance between the user and thedistance detector 120. When the detected distance value D is smaller than the distance threshold T, thedata processing circuit 130 enables the face identification program, as shown inStep 140. - As shown in
Fig. 5 , enabling the face identification program includes that thedata processing circuit 130 determines the luminance corresponding to the detected distance value D and drives the infraredlight transmitters 140 to emit the infrared light according to the corresponding luminance, as shown inStep 141 andStep 142. Then, thedata processing circuit 130 starts the infraredlight camera 150 to start to capture the infrared light image Pir and transmit the infrared light image back, as shown inStep 143. - According to different detected distance values D, the corresponding luminance is determined to emit the infrared light so as to guarantee that the infrared light image Pir can have good image exposure and a clear image capturing effect. The specific implementation is shown in
Fig. 3 , the range of the distance threshold T is divided into the two or more distance intervals D1, D2, D3, the luminance has the plurality of set values I1, I2, I3, and each of the plurality of distance intervals D1, D2, D3 corresponds to one of the plurality of set values I1, I2, I3. - When the detected distance value D is larger than the distance threshold T but still in the distance interval D4 where the infrared
light transmitters 140 are capable of obtaining the detected distance value D, thedata processing circuit 130 does not perform the face identification but is still capable of driving the infraredlight transmitters 140 to emit the infrared light at theset value 14, so that the infraredlight transmitters 140 maintain the set value I4 of the maximum luminance within the capability of the luminance to emit the infrared light, and surroundings are illuminated by the infrared light. - As shown in
Fig. 6 , when the face enters into an image capturing range of the infraredlight camera 150, and the detected distance value D is smaller than the distance threshold T and falls in a range of the distance interval D3, the infraredlight transmitters 140 take the maximum set value I3 as the luminance firstly to emit the infrared light, so that the face can still be fully illuminated at a long distance, and it is guaranteed that the infrared light image Pir is sufficiently exposed and is clear enough. At the moment, if the user continuously approaches theface identification apparatus 100 to have the detected distance value D to fall in the distance interval D2 or the distance interval D1, the luminance of the set value I3 is too high, which results in that the infrared light image Pir is too bright, and consequently, features of the face become unclear. At the moment, thedata processing circuit 130 determines the corresponding luminance as the set value I2 according to the distance interval D2, so that the infrared light image Pir can be moderately exposed and the face is clearly presented in the infrared light image Pir. - As shown in
Fig. 7 , if the face enters a smaller distance interval such as the distance interval D1 at the beginning, the infraredlight transmitters 140 will take the smaller set value I1 as the luminance to emit the infrared light, so that the infrared light image Pir is moderately exposed and the face is clear enough. But if the user moves to change the distance to increase the detected distance value D to be in the distance interval D3 or the distance interval D2, the luminance of the set value I1 will be insufficient, which results in that the infrared light image Pir is underexposed and is too dark, and consequently, the features of the cannot be clearly presented infrared light image Pir. At the moment, thedata processing circuit 130 determines the corresponding luminance to be increased as the set value I3 according to the distance interval D3, so that the face in the infrared light image Pir can be presented clearly. - As shown in
Fig. 5 , at the same time, thedata processing circuit 130 starts the visiblelight camera 160 to capture the visible light image Pv and drives thedisplay panel 170 to display the visible light image Pv, as shown inStep 144. At the moment, the user can regulate the self-position and the posture according to display of thedisplay panel 170 so as to guarantee that the face of the user can be captured by the infraredlight camera 150. - The foregoing infrared
light transmitters 140 may normally emit the infrared light at the maximum luminance such as the set value I4 within the capability to perform infrared light illumination around. The infraredlight camera 150 continuously captures an infrared light image Pir stream and stores the stream to thedata processing circuit 130. But thedata processing circuit 130 at the moment does not perform the face identification and only records a video. As well, the visiblelight camera 160 may also normally capture a visible light image Pv stream. Thedata processing circuit 130 records the video, but does not drive thedisplay panel 170 to display the video. - Then, the
data processing circuit 130 performs the face identification according to the infrared light image Pir and compare whether the face eigenvector combination of the face of the infrared light image Pir matches one of the plurality of preset eigenvector combinations in the eigenface database, as shown inStep 150. The above-mentioned plurality means including two or more. - When the face eigenvector combination matches one of the plurality of preset eigenvector combinations, the face identification is successful, and the
data processing circuit 130 performs the corresponding event, as shown inStep 160. The foregoing corresponding event may be, but is not limited to, door lock opening, attendance recording, access control recording, book borrowing recording, etc. - Thus, by means of the above
face identification apparatus 100 and the face identification method, the face identification program is normally in the idle state, so that hardware of theface identification apparatus 100 does not continuously maintain a high computing load due to continuously performing the face identification program. Thus, a usual load of theface identification apparatus 100 can be reduced, and power consumption is reduced. - In this disclosure, the
data processing circuit 130 may be implemented by hardware or a combination of the hardware and software. For example, thedata processing circuit 130 may further include a system logic chipset, a central processing unit, a non-volatile memory module, a volatile memory module and a graphic processing unit. The non-volatile memory module is configured to store program instructions. The central processing unit may load the instructions from a non-volatile memory to a volatile memory module to be performed so as to start the face identification program and switch the face identification program to be in the idle state. - Foregoing descriptions are only one example of the
data processing circuit 130, not excluding integration of part or all of components into a single integrated circuit component. It is also not excluded that other circuit design is adopted to replace the foregoing components as thedata processing circuit 130. - Referring to
Fig. 8 , aface identification apparatus 100 disclosed by a second embodiment of this disclosure is shown, and is configured to provide an example of anotherdata processing circuit 130. Thedata processing circuit 130 may be divided into three parts including acomputing unit 132, a drivingcircuit 134 of adistance detector 120 and avoltage regulator 136. - The
computing unit 132 includes components such as a central processing unit (CPU) and a system logic chipset and is configured to perform face identification. Thecomputing unit 132 may be connected to an infraredlight camera 150 and a visiblelight camera 160 through a USB bus (or other interfaces such as an MIPI-CSI). Thecomputing unit 132 is connected to thedriving circuit 134 of thedistance detector 120 through an I2C interface (or other interfaces such as an SMBus and a USB). Meanwhile, the drivingcircuit 134 may be connected to thevoltage regulator 136 through two output pins such as general input and output pins GPIO-1 and GPIO-2. The general input and output pins GPIO-1 and GPIO-2 and a base voltage V are in parallel connection with thevoltage regulator 136 respectively through resistors, so that a reference voltage Vref is regulated and output to thevoltage regulator 136 through the general input and output pins GPIO-1 and GPIO-2. Thevoltage regulator 136 is connected to a power source Vdd, such as a power source with +5V voltage, and outputs an output voltage Vout according to the reference voltage Vref to regulate a magnitude of a driving voltage V-LED of infraredlight transmitters 140, so that one of set values I1, I2, I3 and I4 is selected to change a luminance. - The
distance detector 120 may be an IR distance detector, a distance detector and a time of flight (TOF) detector. A detected distance value D detected by thedistance detector 120 is coded by the drivingcircuit 134 and then continuously transmitted back to the CPU of thecomputing unit 132. - When the detected distance value D is larger than a distance threshold T and the detected distance value D is relatively maximum, for example, falling in a distance interval D4 of distance intervals D1, D2, D3 and D4, the
data processing circuit 130 does not perform face identification and does not switch the set values I1, I2, I3 and I4 of the luminance of the infraredlight transmitters 140 either. At the moment, thecomputing unit 132 controls the drivingcircuit 134 of thedistance detector 120 to make the two general input and output pins GPIO-1 and GPIO-2 be both maintained in an open drain state. At the moment, the reference voltage Vref is maintained at the maximum voltage (such as Vref=2.5 V), so that the driving voltage V-LED is output at the maximum voltage preset value (such as 2.5 V). The infraredlight transmitters 140 maintain the set value I4 of the maximum luminance within a capability as the luminance to emit an infrared light, and surroundings are illuminated by the infrared light. - When the detected distance value D is smaller than the distance threshold T and the detected distance value D is relatively small, for example, falling in the distance interval D1 of the distance intervals D1, D2, D3, the
computing unit 132 controls the drivingcircuit 134 of thedistance detector 120 to make the two general input and output pins GPIO-1 and GPIO-2 be both maintained in a low level state. At the moment, the reference voltage Vref is reduced to the minimum value (such as Vref=1.5 V) to make the driving voltage V-LED be reduced to the minimum voltage preset value (such as 1.5 V), so that the infraredlight transmitters 140 maintain the minimum set value I1 as the luminance to emit the infrared light, and an excessively-bright infrared light image Pir is avoided. - When the detected distance value D is smaller than the distance threshold T and the detected distance value D is at an intermediate value, for example, falling in the distance interval D2, the
computing unit 132 controls the drivingcircuit 134 of thedistance detector 120 through I2C, so that the general input and output pin GPIO-1 is switched to be at a low level, and the general input and output pin GPIO-2 is switched to be in an open drain state. At the moment, the reference voltage Vref also falls in an intermediate value (such as Vref=1.833 V) between the maximum value and the minimum value to make the driving voltage V-LED be regulated to an intermediate voltage preset value 1.833 V, so that the infraredlight transmitters 140 maintain the intermediate set value I2 as the luminance to emit the infrared light, and brightness of the infrared light image Pir is moderate and enough to allow thecomputing unit 132 to perform the face identification. - When the detected distance value D is smaller than the distance threshold T and the detected distance value D is relatively large, for example, falling in the distance interval D3, the
computing unit 132 controls the drivingcircuit 134 of thedistance detector 120 through the I2C, so that the general input and output pin GPIO-1 is switched to be in an open drain state, and the general input and output pin GPIO-2 is switched to be at a low level. At the moment, the reference voltage Vref rises to be higher than the intermediate value (such as Vref=2.167 V) to make the driving voltage V-LED rise (for example, rising to 2.167 V), so that the infraredlight transmitters 140 maintain the maximum set value I3 as the luminance to emit the infrared light, the infrared light image Pir with enough brightness of a face at a long distance can also be captured, and the brightness of the infrared light image Pir is enough to allow thecomputing unit 132 to perform the face identification. - In the foregoing one or more embodiments, the three distance intervals D1, D2, D3 corresponding to the three set values I1, I2, I3 of the luminance are only examples, and the corresponding number thereof is not limited to three and may be two or four or more. Meanwhile, at least one distance interval (such as the distance interval D4) can still be set above the distance threshold T, and the distance interval D4 also corresponds to one
set value 14. - In one or more embodiments of this disclosure, the
face identification apparatus 100 switches the face identification program to be in the state of being idle and being awakened and started at any time. Theface identification apparatus 100 determines whether the face identification program is awakened and started according to the detected distance value D. Therefore, theface identification apparatus 100 does not continuously perform the face identification program, so that continuous consumption of hardware computing resources into the face identification program is avoided. Therefore, this disclosure can effectively reduce an average computing load of theface identification apparatus 100 and avoids unnecessary power consumption. Meanwhile, reducing the load of the system can avoid an over-temperature condition of the system to avoid the system entering into a low clock rate operation, so that the face identification program can be quickly completed. - In addition, in at least one embodiment, this disclosure utilizes the detected distance value D to determine the set values I1, I2, I3 of the luminance configured to illuminate, so as to guarantee that the optimal image exposure and image capturing effects of the face can be obtained from far to near, which facilitates performing of the face identification.
Claims (11)
- A face identification method, comprising:performing a distance detection to obtain a detected distance value (D);determining whether the detected distance value (D) is smaller than a distance threshold (T);characterized in thatwhen the detected distance value (D) is smaller than the distance threshold (T), determining a luminance corresponding to the detected distance value (D) and emitting an infrared light according to the luminance;capturing an infrared light image (Pir) and performing face identification to the infrared light image (Pir); andwhen the face identification is successful, performing a corresponding event.
- The face identification method according to claim 1, wherein a range of the distance threshold (T) is divided into a plurality of distance intervals (D1, D2, D3), the luminance has a plurality of set values (I1, I2, I3), and each of the plurality of distance intervals (D1, D2, D3) corresponds to one of the plurality of set values (I1, I2, I3).
- The face identification method according to claim 1, wherein the face identification comprises:analyzing whether there is a face in the infrared light image (Pir), and correcting an image of the face;capturing a face eigenvector combination of the face; andcomparing whether the face eigenvector combination matches one of a plurality of preset eigenvector combinations in an eigenface database.
- The face identification method according to claim 1, further comprising: when the detected distance value (D) is smaller than the distance threshold (T), capturing a visible light image (Pv), and displaying the visible light image (Pv).
- The face identification method according to claim 1, further comprising: when the face identification is successful, displaying a successful identification prompt, wherein the successful identification prompt comprises a user identity data.
- A face identification apparatus (100), comprising:a distance detector (120), configured to obtain a detected distance value (D);an infrared light transmitter (140), configured to emit an infrared light;an infrared light camera (150), configured to capture an infrared light image (Pir); anda data processing circuit (130), configured to determine whether the detected distance value (D) is smaller than a distance threshold (T);characterized in thatwhen the detected distance value (D) is smaller than the distance threshold (T), the data processing circuit (130) determines a luminance corresponding to the detected distance value (D), and drives the infrared light transmitter (140) to emit the infrared light according to the luminance; and the data processing circuit (130) performs face identification to the infrared light image (Pir), and when the face identification is successful, the data processing circuit (130) performs a corresponding event.
- The face identification apparatus (100) according to claim 6, further comprising:a visible light camera (160), configured to capture a visible light image (Pv) and transmit the visible light image (Pv) to the data processing circuit (130); anda display panel (170), configured to be driven by the data processing circuit (130) to display the visible light image (Pv).
- The face identification apparatus (100) according to claim 7, further comprising: a casing (110) comprising an internal space and a front surface (112); the distance detector (120) is disposed on the front surface (112) and configured to obtain the detected distance value (D) in front of the front surface (112); the data processing circuit (130) is disposed in the internal space; and the infrared light transmitter (140), the infrared light camera (150), the visible light camera (160) and the display panel (170) are disposed on the front surface (112).
- The face identification apparatus (100) according to claim 7, wherein when the face identification is successful, the data processing circuit (130) drives the display panel (170) to display a successful identification prompt, wherein the successful identification prompt comprises a user identity data.
- The face identification apparatus (100) according to claim 6, wherein a range of the distance threshold (T) is divided into a plurality of distance intervals (D1, D2, D3), the luminance has a plurality of set values (I1, I2, I3), and each of the plurality of distance intervals (D1, D2, D3) corresponds to one of the plurality of set values (I1, 12, 13).
- The face identification apparatus (100) according to claim 6, wherein the data processing circuit (130) comprises:a computing unit (132) electrically connected to the infrared light camera (150);a driving circuit (134), electrically connected to the computing unit (132) and the distance detector (120), wherein the detected distance value (D) is coded by the driving circuit (134) and transmitted to the computing unit (132); anda voltage regulator (136), connected to the driving circuit (134) and the infrared light transmitter (140), and receiving a reference voltage (Vref) from the driving circuit (134);wherein the computing unit (132) controls the driving circuit (134) to switch the reference voltage (Vref) according to the detected distance value (D), so as to regulate a magnitude of a driving voltage (V-LED) output to the infrared light transmitter (140) by the voltage regulator (136), so that the luminance of the infrared light transmitter (140) corresponds to the detected distance value (D).
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW109107546A TWI777141B (en) | 2020-03-06 | 2020-03-06 | Face identification method and face identification apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3876144A1 true EP3876144A1 (en) | 2021-09-08 |
Family
ID=72322356
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP20193842.0A Pending EP3876144A1 (en) | 2020-03-06 | 2020-09-01 | Face identification method and face identification apparatus |
Country Status (4)
Country | Link |
---|---|
US (1) | US11282296B2 (en) |
EP (1) | EP3876144A1 (en) |
JP (1) | JP7235707B2 (en) |
TW (1) | TWI777141B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108629298A (en) * | 2018-04-23 | 2018-10-09 | Oppo广东移动通信有限公司 | Face identification method, electronic device and non-volatile computer readable storage medium storing program for executing |
EP3425558A1 (en) * | 2017-07-05 | 2019-01-09 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Iris collection method and electronic device |
CN110334665A (en) * | 2019-07-10 | 2019-10-15 | 贵州安防工程技术研究中心有限公司 | A kind of face identification system and method for 3D identification |
CN110463183A (en) * | 2019-06-28 | 2019-11-15 | 深圳市汇顶科技股份有限公司 | Identification device and method |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4380412B2 (en) | 2004-05-10 | 2009-12-09 | 株式会社デンソー | Imaging control apparatus and program |
JP2008181468A (en) | 2006-02-13 | 2008-08-07 | Smart Wireless Kk | Infrared face authentication apparatus, and portable terminal and security apparatus including the same |
TWM364223U (en) * | 2009-04-16 | 2009-09-01 | Hi Sharp Electronics Co Ltd | Light source control device for surveillance camera |
US20160275518A1 (en) * | 2015-03-19 | 2016-09-22 | ecoATM, Inc. | Device recycling systems with facial recognition |
EP3304427A4 (en) * | 2015-06-05 | 2019-06-12 | Seeing Machines Limited | Protective system for infrared light source |
US10169671B2 (en) * | 2017-02-17 | 2019-01-01 | Motorola Mobility Llc | Face detection with temperature and distance validation |
JPWO2019026925A1 (en) | 2017-08-01 | 2020-07-30 | パナソニックIpマネジメント株式会社 | Personal authentication device |
CN108270966A (en) * | 2017-12-27 | 2018-07-10 | 努比亚技术有限公司 | A kind of method, mobile terminal and storage medium for adjusting light filling brightness |
EP3564748A4 (en) * | 2018-02-27 | 2020-04-08 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Control method, control apparatus, terminal, computer device, and storage medium |
CN208873174U (en) * | 2018-10-18 | 2019-05-17 | 汉王科技股份有限公司 | A kind of testimony of a witness veritification device |
-
2020
- 2020-03-06 TW TW109107546A patent/TWI777141B/en active
- 2020-08-19 US US16/997,910 patent/US11282296B2/en active Active
- 2020-09-01 EP EP20193842.0A patent/EP3876144A1/en active Pending
- 2020-10-01 JP JP2020166626A patent/JP7235707B2/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3425558A1 (en) * | 2017-07-05 | 2019-01-09 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Iris collection method and electronic device |
CN108629298A (en) * | 2018-04-23 | 2018-10-09 | Oppo广东移动通信有限公司 | Face identification method, electronic device and non-volatile computer readable storage medium storing program for executing |
CN110463183A (en) * | 2019-06-28 | 2019-11-15 | 深圳市汇顶科技股份有限公司 | Identification device and method |
CN110334665A (en) * | 2019-07-10 | 2019-10-15 | 贵州安防工程技术研究中心有限公司 | A kind of face identification system and method for 3D identification |
Non-Patent Citations (1)
Title |
---|
TURK M A ET AL: "Face recognition using eigenfaces", PROCEEDINGS OF THE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION. LAHAINA, MAUI, HAWAII, JUNE 3 - 6, 1991; [PROCEEDINGS OF THE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION], LOS ALAMITOS, IEEE. COMP., vol. -, 3 June 1991 (1991-06-03), pages 586 - 591, XP010023267, ISBN: 978-0-8186-2148-2, DOI: 10.1109/CVPR.1991.139758 * |
Also Published As
Publication number | Publication date |
---|---|
US11282296B2 (en) | 2022-03-22 |
JP2021140734A (en) | 2021-09-16 |
CN113361308A (en) | 2021-09-07 |
US20210279446A1 (en) | 2021-09-09 |
TWI777141B (en) | 2022-09-11 |
TW202134946A (en) | 2021-09-16 |
JP7235707B2 (en) | 2023-03-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10627887B2 (en) | Face detection circuit | |
US20140050360A1 (en) | Systems and methods for presence detection | |
US9367925B2 (en) | Image detection and processing for building control | |
US8836505B2 (en) | Illumination control system and method for controlling illumination | |
WO2020073572A1 (en) | Control method, control device, depth camera, and electronic device | |
US20140104436A1 (en) | Circuit to adjust backlighting for a display device | |
US20230269462A1 (en) | Wake-up method for surveillance camera, surveillance camera, and non-transitory computer-readable storage medium | |
US20160094763A1 (en) | Asymmetric multiphase boost for display backlights | |
US8351660B2 (en) | Apparatus and method for detecting specific object pattern from image | |
CN109492620A (en) | Monitoring device and its control device, post monitoring method and readable storage medium storing program for executing | |
US9727160B2 (en) | Displacement detection device and operating method thereof | |
Ma et al. | Automatic brightness control of the handheld device display with low illumination | |
US11423762B1 (en) | Providing device power-level notifications | |
EP3876144A1 (en) | Face identification method and face identification apparatus | |
WO2022041220A1 (en) | Electronic device, and response operating method of electronic device | |
CN114424145B (en) | Proximity detection method, terminal and storage medium | |
CN108647647B (en) | Control method and control device of air conditioner and air conditioner | |
US10834366B1 (en) | Audio/video recording and communication doorbell devices with power control circuitry | |
CN113287081A (en) | Electronic equipment control system and method | |
CN113267943B (en) | Method for adjusting brightness of standby indicator light, projector and readable storage medium | |
CN114079709A (en) | Driver mechanism for rolling shutter sensor to obtain structured light pattern | |
CN217061038U (en) | Power supply circuit system and face recognition equipment | |
US20240257495A1 (en) | Training system and training method for human-presence detection model | |
CN118158522A (en) | Method for controlling cruising of imaging device, and storage medium | |
Llorente et al. | Characterizing Illumination Levels of an Image Sensor as Input to a Camera Based Illumination Controller |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20210629 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20221212 |